10-multi-domain-comprehensive.yaml 2.6 KB

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768697071727374757677787980
  1. id: contextscout-multi-domain
  2. name: "ContextScout: Multi-Domain Comprehensive Discovery"
  3. description: |
  4. Tests that ContextScout correctly discovers context files across multiple domains
  5. when asked about a complex task that spans multiple areas.
  6. This validates:
  7. - ContextScout searches across multiple domains
  8. - ContextScout combines relevant files from different areas
  9. - ContextScout prioritizes based on task requirements
  10. - ContextScout doesn't miss critical cross-domain files
  11. Expected Files (in priority order):
  12. 1. .opencode/context/openagents-repo/core-concepts/agents.md (CRITICAL)
  13. 2. .opencode/context/core/standards/code-quality.md (CRITICAL)
  14. 3. .opencode/context/core/standards/test-coverage.md (HIGH)
  15. 4. .opencode/context/openagents-repo/guides/adding-agent.md (HIGH)
  16. 5. .opencode/context/core/standards/documentation.md (MEDIUM)
  17. This test SHOULD FAIL if:
  18. - ContextScout only searches one domain
  19. - ContextScout misses critical files from other domains
  20. - ContextScout doesn't combine findings effectively
  21. category: developer
  22. agent: subagents/core/contextscout
  23. prompts:
  24. - text: |
  25. Search for context files related to: creating a new agent with tests and documentation
  26. Task type: agent-creation, testing, documentation
  27. Search intent: I need to create a complete agent including implementation, tests, and docs
  28. Return:
  29. - Exact file paths to relevant context files
  30. - Brief summary of what each file contains
  31. - Priority order (critical, high, medium)
  32. Focus on:
  33. - Agent creation process
  34. - Code quality standards
  35. - Testing requirements
  36. - Documentation standards
  37. - Best practices across all areas
  38. approvalStrategy:
  39. type: auto-approve
  40. behavior:
  41. forbiddenTools:
  42. - task
  43. - bash
  44. - write # Read-only agent
  45. - edit # Read-only agent
  46. minToolCalls: 1
  47. maxToolCalls: 10
  48. timeout: 90000
  49. tags:
  50. - contextscout
  51. - discovery
  52. - multi-domain
  53. - comprehensive
  54. - critical
  55. # Expected outcome:
  56. # - Uses glob to find files across multiple domains
  57. # - Uses read to load:
  58. # * .opencode/context/openagents-repo/core-concepts/agents.md
  59. # * .opencode/context/core/standards/code-quality.md
  60. # * .opencode/context/core/standards/test-coverage.md
  61. # * .opencode/context/openagents-repo/guides/adding-agent.md
  62. # * .opencode/context/core/standards/documentation.md
  63. # - Returns structured response with:
  64. # * Priority 1 (CRITICAL): agents.md + code-quality.md
  65. # * Priority 2 (HIGH): test-coverage.md + adding-agent.md
  66. # * Priority 3 (MEDIUM): documentation.md